RISCV LLVM benchmark
Nobody has claimed this yet.
- Dominant language
- Shell
- Stars
- 277
- Forks
- 55
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
Recently I ran the coremark on HiFive1 board, https://www.sifive.com/products/hifive1/
With GCC, my test result is 2.66, the official data is 2.73. They are very close.
With LLVM built from https://git.llvm.org/git/llvm.git 02bcd70eea, the test result is 1.79, about 2/3 of the GCC result.
At the same time, I found similar comparison here:
http://www.diva-portal.se/smash/get/diva2:1114115/FULLTEXT02.pdf
In this Thesis, the LLVM dhrystone result is not good.
Are these results reasonable? Do you have internal test result? thanks a lot.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Review the CoreMark results reported for the HiFive1 board, comparing GCC with the LLVM commit 02bcd70eea. Check the linked LLVM repository and the cited Dhrystone thesis for comparable benchmark methodology. Done means establishing whether the performance gap is expected or producing a reproducible discrepancy for maintainers to investigate.
Written by the indexing model from the issue text.
Assessment
- Domain
- compilers, embedded-iot, performance
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100